Neurology -- Correspondence for Engelhart et al., 59 (12) 1915-1921

Correspondence published:

Diet and risk of dementia: Does fat matter?: The Rotterdam Study: William B Grant (11 April 2003)

Does Rotterdam study question the link between fat and dementia?: Alexei R. Koudinov (15 January 2003)

Diet and risk of dementia: Does fat matter?: The Rotterdam Study 11 April 2003

William B Grant
Newport News VA
Send Correspondence to journal:
Re: Diet and risk of dementia: Does fat matter?: The Rotterdam Study

wbgrant{at}infi.net William B Grant

The recent paper on risks associated with dietary fats for AD and vascular dementia by Engelhart et al. [1] is interesting but as indicated in the paper, at odds with a several recent papers. The question is why?
The ecologic study looked at entire populations. [2] It was found that diet 4 years prior to the prevalence data had the highest correlation with AD. Dietary fat was the highest individual risk factor, but including fish in a multiple linear regression improved the correlation. Total energy supply was also highly correlated with AD. The authors of [1] considered fat normalized to total energy. To check on the validity of this approach, the data used in [2] with the addition of more recent data from India [3] were run in regression analyses. The adjusted r2 for total fat is 0.94, F = 179, p<0.001; for normalized fat, the adjusted r2 is 0.90, F = 97, p<0.001. For fat and fish, the adjusted r2 is 0.95, F = 111, p<0.001; for fish and normalized fat, the adjusted r2 is 0.91, F = 57, p<0.001. Thus, it seems inappropriate to use normalized fat in the analysis.
The range of dietary supply of fish oil in [1] was 2.9 g/person/day in Sweden to 0.7 g/person/day in the U.K. This range is greater than in [1]. A recent study in France [4] involving 135 new cases of AD reported a hazard ratio of 0.73 (95% confidence interval 0.52 to 1.03) after accounting for education. Thus, there is support for fish oil reducing the risk of AD. The mechanisms may include reducing inflammation, an important factor in the etiology of AD, as well as providing the n-3 fatty acids important for proper brain development and functioning.
A cohort study in New York [5] found that dietary fat was an important risk factor for AD for those with the APOE e4 allele (hazard ratio of fourth quartile with respect to the first+2.31 (95% C.I. 1.09-4.89), p value for trend=0.02) but not for those without that allele (hazard ratio of fourth to first=1.15 (95% C.I. 0.068- 1.93), p value for trend=0.02).
Comparing the approaches in [1, 2, 4), it appears that [1] normalized fat to energy, omitted those with early onset dementia (those who didn't respond to the invitation to participate), and had a population with low fish consumption. In addition, no genetic analysis was conducted.
References:
1. Engelhart MJ, Geerlings MI, Ruitenberg A, et al. Diet and risk of dementia: Does fat matter? The Rotterdam Study. Neurology 2002;59:1915- 1921.
2. Grant WB. Dietary links to Alzheimer's disease. Alz Dis Rev 1997;2:42-55.
3. Chandra V, Pandav R, Dodge HH, et al. Incidence of Alzheimer's disease in a rural community in India: The Indo-US Study. Neurology 2001;57:985-989.
4. Luchsinger JA, Tang MX, Shea S, Mayeux R. Caloric intake and the risk of Alzheimer disease. Arch Neurol 2002;59:1258-1263.
5. Barberger-Gateau P, Letenneur L, Deschamps V, et al. Fish, meat, and risk of dementia: cohort study. BMJ 2002;325:932-933.

Does Rotterdam study question the link between fat and dementia? 15 January 2003

Alexei R. Koudinov,
neuroscientist and editor
Rus Acad Med Sci, Neurobiol Lipids
Send Correspondence to journal:
Re: Does Rotterdam study question the link between fat and dementia?

koudin{at}med.pfu.edu.ru Alexei R. Koudinov

The article by Engelhart et al. (Neurology. 59: 1915) is the latest in a series of articles by the same group on Alzheimer's disease (AD) using Rotterdam study data. The article sets and tests the hypothesis that dietary "cholesterol increases and n-3 PUFA reduce dementia risk". The authors conclude that the results do not prove the hypothesis. They also describe the study limitations and warn readers that "larger, prospective studies with longer follow-up periods are needed to confirm our findings."
In my view, there are several flaws that limit the significance and question the conclusion of the article. First, The article does not include the same authors' data on the apolipoprotein (apo) E allele variants reported earlier this year in the identical population of the Rotterdam study in another report on a diet and AD (JAMA 2002, 287: 3223; also see the search results for the reports on apoE using the Rotterdam study). Second, the arguments of the authors on the discrepancy between the results of their previous article on the identical subject using the Rotterdam study data (Ann Neurol 1997, 42: 776) and their present report are not clear.
However, the Neurology article does show proof that the involvement of cholesterol and other lipids in neurodegeneration in general, and AD in particular, extends far beyond the simplistic role of cholesterol in amyloid beta generation.¹ The explanation of the role of cholesterol in AD supports the amyloid hypothesis but still does not explain the disease and provide its pathogenetic cure.²
The usage of statins is not clearly understood (or other lipid modulating substances) in treating Alzheimer's which may yield problems rather then a cure. This happened earlier this year during the translation process of non validated amyloid hypothesis into the vaccination clinical application.²
We discussed our vision of the role of brain fat (particularly cholesterol) break as a primary etiologic factor in AD.³ We also cautioned of the over-simplification at the recent International Symposium and at AlzForum live discussion on 'Cholesterol and Alzheimer's disease.' Another warning is provided in three Neurology reports published earlier this year.⁴
Based on the above, I do not think that the Rotterdam study questions the link between fat and dementia risk.
Competing financial interests: none
Sincerely,
Alexei Koudinov, MD, PhD
neuroscientist and editor
http://anzwers.org/free/neurology
http://neurobiologyoflipids.org
References:
1. Koudinov AR, Koudinova NV, Beisiegel U. Cholesterol homeostasis failure at neuromuscular junctions and CNS synapses: a unifying cause of synaptic degeneration ? Neurology online� (2002) [ FullText ]; Koudinov AR, Koudinova NV. Alzheimer�s anti-amyloid vaccination and statins: two approaches, one dogma. The time for change. British Medical Journal online (2002) [ FullText ]; Wolozin B. Cholesterol and Alzheimer's disease. Biochem Soc Trans. 30(4): 525-9 (2002) [ PubMed ].
2. Koudinov AR, Smith MA, Perry G, Koudinova NV. Alzheimer's disease and amyloid beta protein. Science online (2002) [ FullText ];� Koudinov AR. Amyloid was never clearly implicated in Alzheimer's disease, so look at Abeta from a different angle. BMJ online.� (2002) [ FullText ]; Koudinov AR, Koudinova NV. Alzheimer�s anti-amyloid vaccination and statins: two approaches, one dogma. The time for change. BMJ online.� (2002). [ FullText ].
3. Koudinov AR, Koudinova NV. Brain Cholesterol Pathology is the Cause of Alzheimer's Disease. Clin Med Health Res. clinmed200110005 (2001) [ FullText ] [ Companion article ] [ Authors preface ] [ ARF hypothesis ].
4. Fassbender K, et al. Effects of statins on human cerebral cholesterol metabolism and secretion of Alzheimer amyloid peptide. Neurology. 59(8), 1257-8 (2002) [ PubMed ]; Koudinov AR, Berezov TT, Koudinova NV. Cholesterol and Alzheimer's disease: is there a link? Neurology. 58(7), 1135 (2002) [ FullText ]; Gaist D., et al. Statins and risk of polyneuropathy: a case-control study. Neurology. 58(9), 1333-7 (2002) [ PubMed ].